Each bad model that is learned is typically a valid model on some energy/velocity/ mass scale. Newtonian mechanics is a valid for sub-lightspeed velocities. Sol

Each bad model that is learned is typically a valid model on some energy/velocity/ mass scale. Newtonian mechanics is a valid for sub-lightspeed velocities. Solving those problems using full general relativity is possible, but not useful. Modeling molecules as masses connected by springs is valid on short time scales and low energies. You can go on and on. These models are not "bad", they are just approximations that are only valid for a subset of all problems. Without these approximations, you cannot even solve most problems. Its not like you want to teach and understand all of chemistry from a quantum chromodynamics perspective. Perhaps I misunderstood your point, but the history perspective is useful, because each of the models builds upon one another and slowly reveals the stranger and harder to describe parts of nature.