I think there's a place for both...
I see themes as working somewhat like kits did in 2e, but with less attachment to a particular class. (Which I think is awesome.) I liked kits a lot. However, kits had some issues as well. They couldn't seem to decide on whether a kit was your origin story or a sub-sub-class. So you had fighter kits for say..."gladiator", but also for "peasant hero". Nonetheless, they were a nice "something less than a class" tweak.
Which is where Prestige Classes come in. I was so very excited when I read about Prestige Classes before 3e came out. I was still excited when the first couple came out....then they just got worse and worse. Having to carefully pick your feats, skills, and spells from the moment of character creation sorta defeated the purpose of "organic" character growth. That's not to mention all the overspecialized, unbalanced, or just plain weird stuff that they got into (Blood Mage, I'm lookin' at you.) In the end, I ended up almost hating them, and agree with most of the criticisms leveled at them.
4e's multiclassing and paragon paths were one aspect of that system that I didn't really like too much. I'm not so familiar with the themes idea that they came out with later. (My group was done with 4e by that time.)
I would certainly like something akin to these mechanics to be in D&DN, but I can't say I'm so enamored with any of the previous iterations that they couldn't be improved upon or replaced entirely. My biggest criterion for rules nowadays is efficiency, and this is one of those areas that seems to quickly fall into a bloat trap. I don't want to be working up NPCs and be stymied by looking through bunches of Prestige Classes and the like, worried about prereqs. If they were something like secondary themes, that a PC could pick up as they went along, that might work well.
Whatever they do, I hope their criteria for inclusion is the fun that it adds to the game, not how the architecture makes it easy to fill more splatbooks.