Fundamental laws are "just" models, and models are only as good as we can validate them. My day-to-day involves running simulations of physical systems and while things like the Navier-Stokes equations are well known and understood (to the best we can), the difficult part comes in the constants we use in the models to describe specific aspects. The Navier-Stokes equations themselves can describe and tell us how fluids behave, but we still need constitutive models to describe the fluid. We get those models by proposing ideas and testing them.