Campaign research questions are generally on topic as long as they are acceptable questions according to other site critera, though we should suggest asking in more appropriate SEs in cases where it is probably better.
Interestingly, I was a major opponent of campaign research questions in the original discussion, but I have come around and disagree with the majority of the "let's ban them" reasoning. here's the deal. To ban a kind of question as a "site rule," it needs to present a clear and present danger to the site. Anything else should go under normal site voting rules. So let's evaluate.
Are There A Lot Of Them?
These questions are not at all frequent. See @AceCalhoon's survey of the questions, only a handful really touch on this problem directly at all. I personally tolerate way more dumb questions in other categories I don't like, so it doesn't make me too sad.
Are They Outside The Site's Scope?
In other words, are they on topic in the most simple sense of being about the subject matter of the site? (This doesn't mean we allow them in and of itself) - like on Gaming, a shopping question might be "on topic" but banned for other reasons.)
Most of these questions seem to me to be legitimately and directly useful to gamers. More than that, they benefit from an answer from a gaming context.
Here's my personal yardstick. Would you expect to see an article written on this topic in an RPG magazine? If so, it is appropriately on topic. For example, with the pirate-based Skull and Shackles adventure path on, RPG gamers are legitimately interested in how to implement nautical travel in their games. An answer to this question from History.SE is likely to be too rooted in the real world to be helpful. Part of the brilliance of Steve Jackson Games, for example, is in their historical sourcebooks for GURPS. These translate "history" into "history in a form useful for gamers". I remember how impressed I was with their Conan sourcebook for the same reason, it converted the fragmentary history of the Howard stories into something one could coherently use to game in - a not inconsiderable task. That translation is "not nothing..."
The "non-fantasy threats" question is on topic. A real world zoologist would give crap answers to this, from a "I want to put together a generic 'forest random encounter table'" perspective. I have several Dragon Magazine articles on this exact topic. You could possibly make a "big list" argument against this specific question but that's irrelevant to whether its subject matter is on topic.
Saying that a topic that an RPG product specifically covers is off topic for RPG.SE is a questionable line of reasoning.
The "What commodities exist in the 1930s" is a closable question because it's poorly scoped, not because it's off topic. Frankly "How do I find out what commodities are available" is an on topic better scoped question - there's plenty of RPG products out there to help people with that question!
Is There A Better Place For Them?
Here's a daring concept. Let people put their questions where they think they're going to get the best answer. RPG vs History, Ubuntu vs Linux and Unix, etc. If the question is off topic in your estimation, vote to close. 4 people agree with you and away it goes. If it's not off topic, but you think it's "better" answered somewhere else - well, opinions are like assholes, everyone has one. Suggest it to the OP. But I'm not really interested in enforcing "what someone thinks would be better," mod action should generally be for major wrongs not opinion differences. Migrating against an OP's wishes not for "off topic" but for "better" is hostile and should not be accepted practice anywhere on the network.
I don't agree entirely with @JonathanHobbs' conclusion but I do agree with the bit he linked from gamedev.se, especially given as it's worded as advice to the user and not as "we will close or migrate you for daring to ask it here!"
Is There Another Compelling Reason To Ban Them?
An argument for banning shopping questions (especially without tight guidance) is that they generate poor quality answers. Or arguments. I don't see campaign research questions doing that (except for the meta-argument of "Do we allow them," which doesn't count). So I'm having trouble seeing any way in which they are being disruptive or contributing to people breaking the "real" site rules (as codified by the vote/flag reasons).
Given that there aren't many of them, and they are generally on topic and not causing problems, and that the site mechanism exists to close ones that are off topic or causing problems, I see no reason for a "site rule" on this topic. Site rules should be minimized to be a friendly, open community - we should only have rules over and above the general SE rules and the FAQ if such a rule is sorely needed (like sys-rec questions qualifies). I don't see that this does.
RPG.SE Is Not RPG Rules SE Or Fantasy SE
Just a warning about some of the common misapprehensions I'm seeing in this discussion.
The act of creating, running, and playing RPGs is about much more than game rules/mechanics. If you are looking to reduce questions just because they don't touch on game rules, you will not find an ally in me.
Similarly, "non magic setting" does not make things off topic here. D&D is fine and great, but this site is and should be inclusive of all games and settings. I understand that the industry largely wallows in the "D&D Ghetto" but that is not relevant here.