Maybe "simulation" is the wrong word for it, but "process sim" can still work since we seem to agree on that definition.
And even if you don't see how D&D can work as a process sim - and even I will admit that there are a few inconsistencies, because the designers are only human - you should respect that it's worked well enough for myself, and for others, that we have reason to see that it can continue to do so in the future.
I mean, if you treat it as primarily a game, then I can see how it's worked that way for you, and I would be disappointed if they made it unplayable as such.
See, but that's the thing. The claim here is that something changed as we went along. That 4e suddenly made process sim a problem. And that it might be a problem in 5e because of things like Second Wind and short rests and the like. But, I've been using 3e as the example here. I could do the same thing with 2e and 1e as well just as easily.
Passing it off as a "few inconsistencies" is a bit much. When something as basic and fundamental as HP and HD are inconsistent, how can you actually have any consistency in the rest of the game? There is no process to simulate since the process is unknowable. You can't look at any part of the game and say, "Ok, we start at A, pass through B and reach C." which is a fundamental concept of process sim.
Not without ignoring vast swaths of the game. Healing overnight is a problem because it's too fast? But, healing in two nights is consistent? Really? Damage on a miss is inconsistent but not being able to actually tell how and why you missed is? We almost all agree that a miss can make contact with a target - 3e had touch AC's specifically for this idea - so, why not have an ability to allows those attacks to deal minor damage? It's not inconsistent with anything.
The problem I'm having is that people are trying to claim their personal preference as having some sort of objective value. "I don't like DoaM" is perfectly fine. No problems. But, "DoaM is inconsistent with the in game fiction" doesn't hold any water and it's demonstrable that it doesn't hold any water. "Hit points are consistent" is flat out untrue.
See, the whole "D&D as process sim" argument would be a lot more credible if it had appeared at any time BEFORE the edition wars surrounding 4e. After all, if it was simply a play style thing, 4e is hardly the first edition to be pretty bad at process sim. AD&D doesn't do process sim at all and doesn't pretend to. There's a reason we got games like GURPS and Role Master when we did and it's not because AD&D was a great process sim game.
But, the "it worked well enough for me" line is problematic because you cannot actually explain how it works for you. There's so many consistency holes, even using the 3e mechanics, that it's frankly baffling why you would even try. Sure, you could cook an egg on a radiator, but a frying pan would be a lot better and I'd never try to claim that a new radiator was bad because it doesn't fry eggs.