The copy of the real May's brain wouldn't have been able to function in the simulated brain if the latter hadn't been equal in complexity and function to a human brain.
That's not not necessarily so. It might function if the complexity were within some requisite order of magnitude, but not in a way that was completely accurate.
And as a more general remark, we shouldn't conflate what's granted in the fictional world with what's possible in the real world. For example:
With sufficient processing power, a computer model could [be] as complex as, or even more complex than, actual reality.
That's like saying, with sufficiently many rungs, a ladder could reach Andromeda.
Reality spans the
whole universe.
There is no known or even theoretical way of constructing computer memory with the property that the amount of information stored in the memory equals or exceeds the amount of information that represents the memory storage unit itself. Put simply, the storage unit itself is always more than what it is storing. We would run out of universe in which we could build memory before we could build enough memory to represent the portion of the universe we had occupied with memory cells.
Further, it is not theoretically possible to perfectly isolate one part of the universe from the effects of any other.
It is therefore not theoretically possible,
in the real world, to create a computerized simulation of the universe that has perfect fidelity to reality. But again, going back to what I led with, perfect fidelity, given its theoretical impossibility, is not the goal with actual simulation. Creating a model capable of inferring the quantities that the researcher is interested in within some approachable margin of error,
while completely ignoring all others, is the goal.
Throw in some comic book magic, and it's a different ballgame, of course.
