In most introductory discussions on spin-orbit coupling in the hydrogen atom that I've come across, the analysis is often performed on a Bohr-orbital model. The electron is treated as orbiting the nucleus in a circular orbit. They then transform to the frame of the electron, and then assert that the orbital motion can be analyzed as a current loop. But I cannot help but question the validity of such a model. While admittedly the semiclassical Bohr model is able to make some rather accurate predictions of some physical parameters, this particular case feels kind of stretched to me. Could someone perhaps assist me in clearing this doubt up? Thanks.