There is an observation field for linked observations which I often use when a photo is of more than one organism to clarify which the submission represents, or for repeat observations of one organism over time. It seemed like the linked observations were not noticed by identifiers, so I got in the habit of calling out the purpose of a linked observation in the notes.
I also use notes to record anything not captured by the structured data items which I perceive would be of interest to a naturalist (as opposed to the lister or taxonomist): behavior, context, details not captured by image. I don’t worry about whether there is a short-term value to the effort. In the long run, any pertinent natural history detail may have value to somebody someday — but only if it is on the record. For me, the image(s) and structured inputs are sufficient probably ≥95%, but I don’t hesitate when they are not.