And also that other inconvenient fact that neuroscientists barely understand the brain at all.
It might be. You can invert a model by training a neural net to compute its inverse.