Science is never "settled". Instead, there's a consensus that most scientists agree on and build on top of.
Building on top of the consensus usually explains things — until it doesn't. Then they try to tweak that consensus in order to explain stuff that can't be explained otherwise.
Occasionally, you reach a limit in which there's way too many things that don't fit into the consensus that tweaking it is no longer possible. Then, someone proposes a paradigm shift that explains some of the new inconsistencies, and the rest of the scientists reexamine their previous "facts" in order to make them fit with the new consensus.
Let's use Pluto as an example. We were all convinced that there were 9 planets in the Solar System, and counted Pluto as the last one. Then, we've discovered more of Pluto-like planets, and discovered that Pluto is no more special than the rest of them. So, we redefined the term planets, introduced a new term called dwarf planets, and put Pluto into it alongside Ceres (that we knew of before Pluto, but decided to ignore), Haumea, Makemake, and Eris. Now, "must clear its orbit" is a prerequisite for the term "planet", while the objects that don't clear their orbits (but satisfy other rules for a "planet") fit into the term "dwarf planet".
