I agree with your initial argument about why we aren't Boltzmann brains, but I don't think that the same argument follows regarding the simulation hypothesis. A simulation would imply a set of rules being set up, of which your senses, and the reality they experience, are an emergent property. That is, there are rules, and your senses are generally consistent with whatever they are.
With a Boltzmann brain, there's no reason whatsoever to think that any senses are consistent from one moment to the next or that any action can predictably yield any reaction. It would all just be random.
What OP is saying, I think, is that the argument that led us to believe that Boltzmann Brains are more likely than normal brains is itself product of the physics of the world we experience.
But then, if we conclude that we must be a BB, there is no reason for the physics of the universe in which this BB exists to be the same as the physics of the universe we experience. Hence, the argument breaks down, because BB are only very likely in the 'simulated universe', but have no reason to be in the 'real universe' (which has no reason to abide by the same laws of physics).
In other words, the assumption that we are a BB, because they are so much more likely, invalidates the argument that BBs are much more likely.
Whatever the physics of the parent universe are, they either prohibit BBs (and then that universe would be ours, which it presumably isn’t) or are unlikely to affect BBs in a way that disallows creating BBs that observe a similar universe. It’s akin to “why tf I’m a human on Earth 2024 of all planets and things” argument. Well, four reasons: you’re, a human, here, now.
Even if we assume that this universe we observe as BBs is as well the parent universe, it can produce not only human BBs obviously and not only coherent experiences. Complete nonsense is normal here too.
So if a universe allows for BBs (iow, spontaneous temporary observers with state that somehow produces the me-is-observing effect), it allows for all sorts of experiences, and given enough iterations could reconstruct itself and everything imaginable and unimaginable in a sequence of arbitrary length. “You’re certainly not”doesn’t follow here. It may be or it may be not. You’re a human, here, now.
But if you can have a Boltzmann brain, then you can also have stable environment pop into existence. In a simulation, there's no guarantee the simulation follows the parent universe's rules. It could be a simulation with made-up rules. Maybe exploring alternate kinds of universes.
Yet, I observe what appears to be consistency and rules. Randomness happening to manifest in a way that mimics consistency and rules is very very unlikely, probably more unlikely than there actually being consistency and rules.
With a Boltzmann brain, there's no reason whatsoever to think that any senses are consistent from one moment to the next or that any action can predictably yield any reaction. It would all just be random.