The microprestress-solidification theory of concrete creep is first recast in a different but equivalent format, which permits elimination of one model parameter without affecting generality of the model and can serve as the basis of an efficient numerical method. The mechanical model is then combined with a moisture transport model and used in finite element simulations of shrinkage and creep of slabs and prisms. Comparison to experimental data reveals that the model can provide good fits of some of thecreep curves from the literature but fails to properly reproduce the experimentally observed size effect on drying creep. The main reason is that the originally postulated equation for microprestress relaxation is too simple and does not cover a full spectrum of relaxation times. This leads to a delay between the humidity changes and the resulting increase of viscosity that contributes to drying creep. A modification which takes into account the instantaneous effects on viscosity by an additional viscousdashpot is outlined and the resulting improvement of the model performance is demonstrated.