I would argue that your first and third points are not very strong.

I think that it is not useful to protect an idea so that it is only presented in its 'cool' form. A lot of harm is done by people presenting good ideas badly, and we don't want to do any active harm, but at the same time, the more ways and the more times that an idea is adequately expressed, the more likely that idea will be remembered and understood.

People who are not used to thinking in strict terms are more likely to be receptive to intuition pumps and frequent reminders of the framework (evidence based everything). Getting people into the right mindset is half the battle.

I do however, agree with your second point, strongly. It is very hard to get people to actually care about evidence, and most people would not click through to formal studies; even fewer would read them. Those who would read them are probably motivated enough to Google for information themselves. But actually checking the evidence is so central to rationality that we should always remind new potential rationalists that claims are based on strong research. If clickbait sites are prone to edit out that sort of reference, we should link to articles that are more reader friendly but do cite (and if possible, link to) supporting studies. This sort of link is triple plus good: it means that the reader can see the idea in another writer's words; it introduces them to a new, less clickbaity site that is likely to be good for future reading; and, of course, it gives access to sources.

I think that one function that future articles of this sort should focus on as a central goal is to subtly introduce readers to more and better sites for more and better reading. However, the primary goal should remain as an intro level introduction to useful concepts, and intro level means, unfortunately, presenting these ideas in weakened forms.