That sounds like more of a criticism of large PRs and long feedback cycles than of code review itself.