Saying that quantum effects are present and saying that the human brain is a quantum computer (or more?) and will get exponential speedup are two very different things.

We have a pretty good understanding of the individual neuron, and we modeled it with good accuracy. We used a very simplified model of it to build deep learning. The problem is the absurd amount of neurons and interconnections that we would have to run to simulate a human brain.

At some point we may find some phenomena for learning that involves some quantum event, but I doubt we have ever to simulate them at the quantum level.