Resources are finite. A "depth-first" search is a poor choice because it takes a lot of resources fully test any given "cure". A "breadth-first" search (as you

Resources are finite. A "depth-first" search is a poor choice because it takes a lot of resources fully test any given "cure". A "breadth-first" search (as you are suggesting) is a poor choice because there are simply too many things to try. It has to be something in between, and researchers use heuristics to decide which branches to pursue. Sometimes they focus too much on the wrong branch.