yes! the additive model works when you are looking for marginal effects across lots of different genetic, environmental, (sub-)phenotypic contexts in a population cross section. other models may not yield such marginal effects but that doesnt mean they aren't reevant at the individual level
