Modeling color print reproduction is difficult, mainly because of light scattering, causing optical dot gain. Most available models are based on macroscopic color measurements, the average value over an area that is large relative to the halftone dot size. The aim of this study is to go beyond the macroscopic approach, to study color print reproduction on a micro-scale level. An experimental imaging system, combining the accuracy of color measurement instruments with a high spatial resolution, opens up new possibilities to study and model color print reproduction. The main focus is to study how the reflectance values of the printed dots and the paper between them vary with the dot area fraction. A previously proposed expansion of the Murray-Davies model is further developed to handle color prints, predicting tristimulus values. The color of the halftone dots and the paper between them is derived from 3D color histograms in CIEXYZ color space. The prediction errors of the model were found to be equivalent, or better, to that of the Yule-Nielsen model using an optimal n-factor. However, unlike Yule-Nielsen, the expanded Murray-Davies model takes into account the varying reflectance of the ink and paper, and preserves the linear additivity of reflectance, thus providing a better physical description of optical dot gain in color reproduction.