how to create an appropriate prompt to result in a specific image in"text to Image"

I have been unable to achieve what I think should be relatively easy. Specifically, I was trying to bring up a picture of Goldilocks (from the Goldilocks and the Three Bears children's story). Many images of her exist in the public domain, as noted in wikimedia commons. FWIW, I can find some good images of Goldilocks in Adobe Stock; and also there are images in Adobe Stock, when I use the search term "Goldilocks", that seemingly have no relationship to her; (e.g., pictures/images of flowers or moons/planets, a fish, cookies.)

However, any time I put the name Goldilocks in the Adobe Firefly prompt, the results are not at all what one would expect (alot of times, just yields nonspecific images/pictures of random children).

Is there a way in the prompting method, to specify or direct the process to a certain image. I am hoping in the final version of firefly I will be able to upload an image (e.g. in Photoshop, Illustrator, etc.) and then have the AI process do the modification (e.g. goldilocks dancing with the 3 bears in an amish barn dance- or some other ridiculous image).

I have gotten similarly poor results with text to image generator when specifying other characters, or historical figures (winnie the pooh [it is in public domain], Isaac Newton, Benjamin Franklin, etc.), which again are appropriately shown in Adobe Stock.

This is confusing, because I thought that Adobe Stock was one of the image databases that Firely draws from.

Perhaps, these tips are in a "getting started with Adobe Firefly" guideline, or in a tip section on "optimizing your prompt for desired results", but for me right know it is just a big roll of the dice.

Thanks.