now if we reduce taxonomy to only the differences between species this would be like discussing the leaves of a tree and ignoring the brances. there are many cases where neighboring leaves come from fundamentally different brances.
Well, yes, but that's why we have phylogenetics.
Phylogenetics has proven to be the best way of inferring the existence of the branches. Taxonomy, on the other hand, is notoriously unreliable as an indicator of genuine branch positions in the tree of life.
the taxonomy of mammals, for example, is now being forced to change because it faces the challenge of molecular phylogenetic insights that conflict with the traditional classification. Taxonomy of mammals prior to molecular phylogeny was simply incorrect. (for example see
Page Not Found | California Academy of Sciences). That is why taxonomy isn't very useful when it comes to inferring deep branches of the tree of life.