Quantum effects can be ignored in transistors down to a certain size, that is, above the feature size cutoff, starting from maxwell's equations we can predict the transistor's behavior mathematically among all the transistor's variables we are concerned with to the degree of precision we want. Below the cutoff, quantum effects have a non-negligible impact on a number of variables we do care about and Maxwell's equations are not good enough for our purposes.

Similarly we may or may not need quantum mechanics to describe consciousness to an extent we are satisfied with. If it turns out our brains store data in quantum states, for example (admittedly I know very little about QM), then sure, QM will have to be into incorporated into an adequate description of consciousness.

I'm not equating transistors or logic gates to neurons, by the way. I'm claiming that in a universe with no apparent absolutes, pinning down assumptions through approximate models is all we can do, that this is useful, and that less sophisticated models, even those that have been superseded by models that work in a broader range of cases, can be adequate for our purposes.

you are the media you consume.