The apsidal motion test in eclipsing binaries is one of the few observational procedures available to study the stellar interior. A careful analysis of empirical data and theoretical models has been performed since more than ten years ago by our group, and in the present communication we show the most recent results obtained from the confrontation between theory and observations.We have carried out an extensive comparison between new stellar models (Claret & Giménez 1992) making use of the latest opacity library by Rogers & Iglesias (1992) and the best available data from double-lined eclipsing binaries (Andersen, 1991). The level at which we can theoretically reproduce the observations of classical absolute dimensions, namely, R and Teff for a given mass, is shown in Figure 1, while Figure 2 shows the fulfillment of the requirements of a common age for the two components of the same system. A very similar good agreement is obtained when we take into account a moderate amount of convective core overshooting. Small differences can be easily explained in terms of metal abundance variations with respect to the adopted average value Z = 0.02. A more detailed check is shown in Figure 3 for some selected systems.