On Better Mental Representations Part I: Adopting 'Thinking Tools' — LessWrong