I often hear the argument that Kepler's Laws/ Classical Mechanics should just not be applied to large scale structures like galaxies since it clearly does not accurately predict their behavior. I can understand how this is an intuitive decision to arrive at, but I also think its unfortunately non-scientific. There is precedent for new theories replacing older ones in special cases, like relativistic effects being important when objects are traveling near the speed of light. However, you could apply relativistic equations to events that are not near the speed of light, and you would still get the right answers, the relativistic effects would just be negligible. In that same logic: if you wanted to Kepler's Laws to NOT apply to galaxies, you would have to re-write them so that their effects were just negligible on large scales, or at large masses. I don't think this is the correct approach, I think Dark Matter is the real solution to the problem, but if you want to suggest changing the law instead, it would have to be a universal change, and that would be even more problematic them detecting matter that doesn't give off any light.