As you pointed out in your footnote, science is a bad topic for journalism, intrinsically. Something like Scientific American is about as good as it gets for lay audiences - solicited longform articles from the relevant experts, meant for a general audience.
[0] The impetus for most articles is usually the publication of a new paper, but this always seems weird because it’s really more of a hook: papers are almost always incremental progress on a problem and most of the article ends up being background and context anyway.