I understand what normal mapping is, but under what circumstances would you want to use it?
My guess is that you would want to use it for less important objects in a scene or in a game engine scenario (where frame rate is everything). I am assuming you would not want to use it in a non-game engine scenario for an object that is a central point of focus.
You might want to use it on the hero object of your scene - I see no problems with this. There are no rules here. You should use a normal map whenever you want and whenever it works for what you need and of course whenever you are able to make one. It’s used in games to imitate denser geometry sometimes even form, yes, but this is not it’s only use. It’s a tool. You can do whatever you wish with it. Normal maps are for small surface detail and they look better than bump maps. What is the use for this? How about whenever you have a surface that is not smooth? Pretty much everything around you when you need to show surface details. Think about cloth, wood, scratches, brushed metal, paint - I just realized it would be a very long list if I continued… It is most definitely relevant in whatever renderer you can think of. Think about surface detail, not it’s form.
I use whatever, bumpmap or normal map doesn’t matter. Bumpmaps are easier to mix correctly, but normal maps doesn’t usually demand high bit depths, especially for large surface area at a slope. A normal map can define such a slope using a single color, whereas a bumpmap would need higher resolution and bit depth to show as a gradient without steps.