The other fields are attempting to describe reality. While Newtonian physics is useful, as an approximation, it's also quite clearly wrong. You can imagine a universe which follows those rules but it's not this universe, and that's why it's wrong. Mathematics doesn't care about this universe, so you can pick whatever rules you want. Imaginary numbers are not "more accurate", they don't invalidate any previous understanding. They are an imaginary concept with interesting properties. For mathematics, that's enough.