I know what you are trying to get at, but it doesn't work that way. The level of estimation should be based off the logic required, meaning one piece of logic could be handled in one module or four. In addition, the effort required to complete a project will be based off the architecture of the model. If you have bad architecture, not following best practices, etc, then the project will be extended due to workarounds that are unforeseen at the earlier stages.
I would advise utilizing the Anaplan Way and going through the Planning Poker exercise where people go through the user stories and decide how hard/easy it will be to complete that user story.
If you are familiar with the "Planning Poker" methodology used in The Anaplan Way, then that should help you. if not, then you need to be familiar with it in general. Check here
A good first step is to divide each module (or actually each user story) according to the screenshot below
Focus on estimating the time needed for a Meuim- Medium module. If based on the skill and knowledge of the team this user story will take 10 hours, then you can calculate how long it'll take each user story/module based on the Base Line chart above. This way, you need to think about how much time ONE type of user story will take and estimate the rest relative to that.
Einas "Give a Man a Fish, and You Feed Him for a Day. Teach a Man To Fish, and You Feed Him for a Lifetime"