Why are scientific models so useful even though they don't correspond exactly to reality?

Sometimes they are useful to have a physical understanding of process, even if it is not entirely accurate. Quantum chemistry is all approximations, however the predictions are sometimes very accurate and this can give us an idea of which molecular orbitals interact...and so on.

Sometimes reality is just to complex to have completely accurate models.

Scientific models, if they are reasonably close to predicting an outcome of reality, are good because to have a close idea of an outcome is far, far better than having no idea at all.

Take hurricane predictions, for example. It is better to know, say, within 100 miles of where the hurricane will hit than have no idea at all. Of course, this saves billions of dollars...you'd never know it they way some people disregard the warnings, but many people do, hence the relatively few deaths we have now compared to 50 years ago.

I think it's because scientific models are developed based on reality and serve to order our thoughts into someting that can be articulated to others, even if they deviate slightly than the original model. They're forms that we can pour though into.

Because they behave the rules of mathematics and thereby allow math models to replace reality to a degree close enough to get useful, if not perfectly accurate, answers.