I don't think we need to revise the policy.
As a bit of background, I had similar questions about the requirement when I first started participating here. But Carey was right then, and is still right for today.
1. Complicated exceptions make it hard for people to know what's expected
Asking people to show an extremely basic attempt at research is straight forward. Anyone can describe the two words they typed into their favorite search engine. If we start having to judge what is "too technical", that will just lead to more disagreements, more rules lawyering, and more posting on Meta. I also don't think having a secret phrase that people have to include to keep their question from being closed is the right plan.
2. The research effort ask is very low
We aren't asking people to perform an exhaustive literature review on PubMed. We're asking them to spend a few seconds to look into the question for themselves using whatever resource they can think of. For example, the question author could have indicated they typed the following into a search engine:
Why does phenylephrine have hydrogen bromide?
Side note: the third hit on Google from the US seems to answer the question
I don't think that is too much to ask. Based on some of the grumpy responses we receive, I agree with you that it may be frustrating for some users. However, I suspect it is also frustrating for users to have to read dozens of trivial or incomprehensible questions.
3. The benefit of the research requirement outweighs the risks
People often underestimate how helpful the research attempt can be at understanding the question author's position. The extra information conveyed by the research attempt is at times the magic that allows an answer author to answer the question. If a question can trivially be answered in the first page of search engine results, it's often extremely difficult to know what the question author is misunderstanding (sometimes due to the Curse of Knowledge). Otherwise, questions without research often need details or clarity or are too broad. You'll just be trading out one close reason for another.
4. We community moderators have a light touch with this close reason
In general, we unilaterally close questions requesting personal medical advice because of ethical considerations or off-topic questions that may cause other harm. We will sometimes close questions that can be trivially answered in the first page of search engine results after giving the question author ample time to edit their question.