RUMBLETiGER: That's sort-of the inverse of the concept of a theme. It looks like you took a build and then came up with a fluff description of what the character actually does, which is fine, and important as it shows the difference between theoretical optimisation (which is just about power) and character optimisation (which is about making a strong character and then being able to justify the build with consistent fluff). Making a build to a theme means you describe what a character can do first, and then come up with stats that can represent it (and hopefully making a build that doesn't totally suck in the process).
StreamOfTheSky: I said 'character concept', not 'accurate replica of another character from another system, setting and medium including all their backstory, notable feats and events in their life'. If you're trying to replicate a specific character from a specific setting, you don't do it with D&D class levels, you give them custom HD, custom templates or custom races to represent their unique abilities, or you recreate the entire setting (and then obviously introduce a number of new classes to reflect the abilities of characters within that setting). It has nothing to do with the melee-vs-casters issue, and everything to do with assumptions the D&D system makes about how characters gain power.
Triple-gestalting won't fix this problem, it'll make it worse. Sure, you might be able to cram abilities into fewer class levels to represent characters with varied abilities without making them high-Epic in power for no reason, but you'll end up with characters who can inevitably do a bunch of things that are totally unrelated to the concept you're trying to use.
Trying to replicate a specific character with class levels will always end up with characters who can do a whole bunch of extraneous crap that the original couldn't do because classes are all packaged up. Skill points are the biggest thing. A 20th level Fighter with 13 int (Say, he wanted to qualify for Combat Expertise) still has 69 skill points, where a 2nd level Expert with 18 Int has a mere 50. This means that even spreading his skills around, a 20th level Fighter could very well be a better blacksmith than most actual blacksmiths (An 18 int 2nd level Expert is actually significantly better than most people in their trade) merely by dumping some excess skill points into Craft. And this is just a Fighter who only has an Int bonus at all because he needs it for a Feat. A high-int Rogue at 20th level's never going to show up in sensible fiction because they're too damn good at too many things.
This is compounded by the fact that D&D characters with class levels in PC classes are expected to have piles of magical items if they're to remain competitive, but other fiction rarely features characters carrying the amount of wealth a PC of their approximate ECL would have.
And then there's the ability scores issue. A high level Sorcerer is likely to have a Cha score of (18 point buy +5 inherent +6 enhancement +5 HD) at least 36, but basically nobody depicted in fiction has Cha 36, even if they're supposed to be Sorcerer-ish characters, or have power far in excess of what you'd expect from an epic level Sorcerer. Cha 36 is a lot; it means you're better at Diplomacy than most actual diplomats, even if you've never taken a single lesson in Diplomacy. You just say 'Hey. I'm awesome.' and people do what you want. Take Garnet from FFIX. She's depicted as summoning creatures capable of destroying entire cities in one shot. To do this, having a primary casting stat of 36 is going to be wholly insufficient, but nothing she did suggested the superhuman intellect or charisma you'd expect from a caster of her power, nor does she possess the selection of skills and feats you'd expect from someone with, say, 72 levels of a casting class.
Anyway, this being said, nothing you mentioned suggests that Luca Blight needs to be a triple-gestalt. Massacring people by the thousand isn't too hard for a low-epic melee type. There's a reason why in my Epic games armies don't exist, and anyone with any followers they care about will have contingencies put in place to pull them out before the Prismatic Dragons and Infernals start duking it out, since all low-level characters do in an Epic fight is be Great Cleave fodder (and due to various abilities that power up when you kill things, may be an active liability to their own allies). Wolf Pack Tactics + Quicksilver Motion + Great Cleave + Epic Speed vs a dense formation of 1st and 2nd level Warriors (i.e., most regular troops) = about 200 dead soldiers in six seconds, unless you get unlucky with natural 1s early on. Being able to shoot fire from your sword sounds pretty Desert Wind. Maybe he'd need a few custom manoeuvres. Surviving a few dozen flaming arrows is nothing (Adamantine Full Plate grants practical immunity to arrows from low-level attackers, cutting the average damage to a bit under 2 per arrow, assuming they can hit at all. The flame adds one point of damage per hit. 48 of those is a mere 142 damage, when a 25th level Warblade with no particular emphasis on Con will have over 300 hitpoints. Someone with the Fast Healing feat will heal that much damage in under five minutes.) I'm skeptical of your claims of 'near-Wolverine' regeneration. Wolverine's ridiculous. He got better from being hit by a nuke once. In any case, unless he actually got pulped and came back to life, a few copies of Fast Healing is probably enough. If he's not significantly above 25th level, he'll probably be too feat-starved to get it, so he might just have to get a few copies of it as bonus feats.
There's not much you mentioned that even suggests he needs to even multi class. Maybe a Fighter/Warblade gestalt to keep his level down while giving him enough feats to take all the combat abilities he needs. The problem is that actually a 25th level character can do so many things that a JRPG boss cannot. You can make a character significantly better than that using the D&D rules. The trouble is that D&D is one of the few settings where superhuman combat ability is by default tied to being really good as a bunch of things completely unrelated to combat, both through the number of skill points the game hands out at high level, the availability of magical items, and the fact that D&D utility spells are usually miles better than blasting spells of the same level, when most high-powered settings usually have things the other way around.
). Your army FINALLY defeats him only after:
