Neural coding is not a solved problem. Rate codes are not sufficient to explain all experimental data. Neither is learning; we know some things, sure, we have

Neural coding is not a solved problem. Rate codes are not sufficient to explain all experimental data. Neither is learning; we know some things, sure, we have models for long- and short-term plasticity, but we have no idea how learning actually works. With computation, there are many hypothesis and dendrites do indeed do more than you would think, but again, these are just small details.