Using Rasch estimates in models

Question

I am just starting to learn about Rasch analysis, but there is something I am a little confused on. So let's say you conduct an analysis using a rating scale model for some Likert scale and everything looks good. When you want to use that scale for a study, let's say as a predictor variable, can you just put the Rasch estimates into a standard regression? Meaning, would you just put the Rasch estimates in logits as the predictor in a standard regression. Or are you then limited to a Rasch regression? Relatedly, does it make sense to use the raw scores from the scale, summed or averaged, if in the Rasch analysis there appear to be equal sized thresholds? I understand the scale creation aspect of Rasch analysis, like under the rating scale model, but I am confused as to how to use the estimates for modeling some DV. Any help is appreciated.