From page 6 with figure 1, to page 10 with figure 3, it's clear that even just changing to structured proofs over prose proofs can be helpful in catching errors. Of course the rest of the paper goes into detail about going even further... I don't know if Lamport's ideas have gained any more acceptance in the mathematics communities, but I'm doubtful. At least if memory serves the ABC papers were classic prose-style...