In order to predict the influence of silvicultural practices on knottiness and canopy structure, models were constructed to predict branch characteristics of Corsican Pine (Pinus nigra subsp. laricio) using both branch and whole tree measurements. Thirty-six trees were sampled in central France from a broad range of age (19- to 71-years-old), current stand density (200 to 2190 trees/ha), dominant height (9.2 to 30.0 m) and site index. Mixed models were fitted on branch length and on branch angle after a linearization procedure. A segmented second order polynomial model was chosen for the diameter profile. Nevertheless, this model appeared to be difficult to be transformed for fitting mixed models; a global model was, however, satisfactory. For realistic simulations with these models, simulated correlated residuals were generated of angle and diameter in order to maintain the negative link between angle and diameter of a branch. The mixed model method provided an improvement of the fit by accounting for inter-tree differences and/or intra-tree similarities; it also improved the simulation. The possibility of fitting mixed-effect non-linear models allows a greater choice in the number of models.