Yes, iterative improvement is the goal. But the problem is when everyone in the field believes X and must not challenge it else lost standing etc., then people don't actually delve into assumptions as much they should. A clip on the topic (from several years back) that I thought stated it well: