I get results that do not fit the query.

Hello @defaultuou5n4zi3zg6,

Thank you for your message. I am sorry you are having this problem.

Your prompt is too long, too conversational, like a story. The model will not understand it well.  The most important object should be as far left as possible and everything related to that object should be grouped together.

In your case, the primary objects are the father and daughter. You should not use sentences to describe them "father is overweight".  You should use "[Qualifiers...] Object [Prepositional phases to describe]".

In this case, I would use something like:

overweight photographer father wearing baseball cap with underweight photographer daughter with long black hair holding camera; with white and yellow bird on head; birds in air

This still might be too complicated for the daughter and you might have to generate something simpler in text to image and then go to generative fill to add final touches, like the white and yellow birds.

I recommend this document: Writing effective prompts to further help you in crafting good prompts.

Thanks,
    droopy

Some things to consider:

I say that the father is overweight and the daughter is of normal weight what is "normal"?

I say that the father is beardless, but the father has a beard. Beardless is not a good phrase try "clean shaven" or "smooth face"

Try this:

A father and daughter taking bird photos in the spring. The father has black hair, smooth face, heavier build wearing a baseball cap. The daughter has black long hair. Both are holding cameras. A small white and yellow bird is on the daughters head. Backdrop is a wooded forest with green leaves

You can also use one of the previous generated images as a style reference to get something like this. Top right seems fairly close.