What? It's not "coincidentally" correct, and yes, it absolutely is a correct model for phenomena at a certain scale (in fact, for the vast majority of phenomena in engineering disciplines for instance).
It's really not a correct model even at the scale of the phenomena for which it is correct, that's why it can't explain what's beyond its scale, nor can anyone draw a line to clearly define the boundaries of the scale in which it is correct. I'm afraid I don't have a more profound argument than that. The math of Newtonian physics merely coincides with what's actually going on in reality, but what's actually going on in reality is as explained by Einsteinian physics. We're only keeping Newtonian physics around because it's close enough as an approximation of natural phenomenon and therefore it is pragmatic, but pragmatic does not mean correct and it remains an estimation nonetheless. It's not the correct model of what's actually going on.
