The nature of physical theory

Physical theories are usually developed via indirect paths with lots of dead ends. An inductive leap is usually involved. You must take what you know and then put it together in a way that seems to make sense and then see if it can predict the results of experiments in the real world.

No physical theory is ever considered to be ultimate truth. Because of this, they can not be proven to be true. They can only be disproven. Physical theories are tools we want to use for as many circumstances as possible. We never want to not have a tool to work with. So when we come across a situation where the tool no longer works (aka able to be used to make a prediction), we must develop a new tool to use.

Once we find a situation where the tool no longer works, we have found the limits of where that theory can be applied. This is known as the theory's range of validity. For example, Newton's theory of gravitation has a narrower range of validity than Einstein's. Another example is Galileo's theory is only correct when there is no other force to cause friction when objects are falling (aka it only works in a vacuum).