Multinomial regression - Interaction with too many categories

Hi all,

I am trying to predict the type of parenting style of an individual using a multinomial regression model. In the model, I include an interaction between two nominal variables, but the output is difficult to interpret due to the many categories in both variables.
My interaction is between the country of origin (v1) and the migrant status (v2). The country of origin has 10 categories, while the migrant status has 3 categories (native, first generations, second generations).

Would make any sense to make a regression by groups (ex. by first generations)? I also saw someone recoding into dummies, but I don't understand what would be the improvement.
Is there any good practice or article that you can suggest?
Any suggestion is welcome. I hope you've had a good 2021 start!