Downloaded URL lists could be used to download media directly from the Tumblr CDN, without needing to parse a blog. This could even happen after a blog goes offline, which would be potentially useful for quickly archiving many blogs, and engaging the more laborious task of downloading images after they go offline.
Handling such URLs in a post-download pass could also allow for enhanced image deduplication. That is a separate feature request, though.