Abstract
The assumptions made in the Gordon-Kim model for the forces between closed-shell systems are stated and their validity and limitations discussed. The model is compared with other interaction potential theories. The various modifications of the original model are classified and a more detailed discussion than hitherto of the reason and method for modifying the model is given. Two classes of modifications give a more reasonable description than the others. Two simplified models within one of these two classes are introduced here. Finally numerical comparisons with the best available data on the homonuclear inert-gas dimers are presented which demonstrate the accuracy one can obtain with this simple and computationally fast model.