Whether it makes more sense to edit an old question (as Lawrence recommends) or to write a new one (as 1006a suggests) depends on the particular circumstances involved.
If the old question is in danger of closure and your additions are true to its original intent (which is not as difficult to ascertain as some EL&U participants seem to think), I believe that bringing the question up to current standards does everyone involved—and the site as a whole—a service.
If the old question is not in danger of closure, I favor answering it as is, and trusting in the good sense of EL&U site participants not to close a perfectly good question by retroactively imposing anachronistic standards to it. My impression is that people on this site do not ignore new answers to old questions—and good answers receive a surprising number of upvotes over time. The many Necromancer badges (1,659, at this writing) that site participants have received for "answer[ing] a question more than 60 days later with score of 5 or more" support that impression.
A further consideration is the benefit to readers of consolidating answers to the same question in a single place. Anytime an old question has already attracted some reasonably good answers (which is most of the time), answering the old question on the original page is a major convenience for future site visitors, because it keeps the good answers together in one place instead of scattering them across multiple pages. Bear in mind that questions about usage and word or phrase origins often draw multiple useful answers—it's one of the things that distinguishes our subject from subjects in which a given question tends to have only one correct answer.
If I want to answer a question that is tangentially related to the original question but is not closely enough connected to it to justify parking my answer there, I either post a new question that focuses on the issue of interest to me or I give up on my planned answer. Whether the connection between the question I want to investigate and the question that actually exists is sufficiently close to bear critical scrutiny is a judgment call, of course, and it's tempting—very tempting—to try to shoehorn intriguing information into a question that doesn't exactly ask for it. Still, at some point, even I have to admit that material I've dug up belongs somewhere else, or nowhere at all.