The incorporation of fresh organic matter (plant residues, manures..) to soil has generally strong, while transient, positive effects on aggregate stability. However, the relationships between the amount and quality of the added organic matter on one hand and aggregate stability evolution on the other hand are poorly known. Such relationships are needed to better manage organic matter in order to improve soil properties. Furthermore, very few studies attempted to develop models able to predict change of aggregate stability due to organic matter inputs to soil. Here, we investigated the specific effect of the rate of addition of plant residues. We added ground maize straw at 2.5, 5, 10, 15 and 20 g C kg-1 to 3-5 mm aggregates from an unstable silty cultivated soil. The samples were incubated for 8 months at 0.01 MPa, monitoring their respiration. They were destructively sampled at regular intervals, to measure their microbial biomass, fungal biomass and aggregate stability. The addition of organic matter induced a transient increase of aggregate stability, with a similar temporal pattern for all doses. Aggregate stability increases were linearly related with the rate of addition of straw at different dates and were highly correlated with total microbial biomass and ergosterol content. We satisfactorily modeled the decomposition of the organic residue using the "CANTIS" model. In parallel, we derived regression coefficients between two biological variables affected by straw addition and easily predicted with CANTIS, i.e. CO2 production and microbial biomass, and aggregate stability. We propose a new model that results from the coupling of a decomposition model, CANTIS, with an aggregate stability function. This new model predicts well the transient increases in aggregate stability after straw inputs to soils as well as the subsequent decreases.