While preprint archives certainly remove one of the major reasons that I used to submit technical reports, sometimes I find there's just no other appropriate title for a work.
To me, the whole notion of a preprint assumes intention to eventually publish in a journal, using a journal-style format and style of argumentation. There are a number of things that I write, however, that are worth archiving in citable form that just are not that.
In my own work, for example, I often end up producing formal project reports for funding agencies. These may incorporate material from papers or that will later be extracted for papers, but the report itself can be a valuable document, especially for talking about implementation details or negative results.
Likewise, highly technical documents such as specifications, architecture documents, standards, best practices recommendations, etc., can be extremely valuable, but are often not appropriate to publish as journal articles. One often also writes journal articles about such things, but there is a big difference between an article explaining the core concepts and value in a standard versus the full grinding detail of its complete specification.
I'm sure there are other examples as well, but the bottom line is this: "technical report" remains a nice catch-all name for any archival technical artifact that's a written document but not a journal article.
