There has been a great deal of interest in estimation of terrestrial biophysical parameters such as vegetation with remotely sensed data. Quantitative estimation of vegetation properties with existing algorithms has been based on empirical relationships established by simple regression. The problem in applying these empirical relationships is that those coefficients proposed vary with vegetation type. To investigate the possible development of an algorithm to quantitatively estimate vegetation properties independent of vegetation type, a model-to-model approach is proposed. This approach first inverts a simple bidirectional reflectance distribution function (BRDF) model with limited data points and simulates multidirectional data. The simulated data are then used in the inversion of a physically based BRDF model to estimate vegetation optical properties (leaf reflectance and transmittance) and leaf area index (LAI). This approach is validated with data. collected from three experiments conducted in cotton, alfalfa, wheat, and pecan fields. A sensitivity analysis and demonstration with multitemporal remote sensing data were performed, and the results show that estimated LAI values agree well with field observations and there is a potential in applying this approach on an operational basis in practice with multitemporal remote sensing data.