Shadow-bands are a low cost alternative when a precision solar tracker is not available. Adequate precision may be achieved if the measured diffuse irradiance is corrected to account for the sky portion blocked by the shadow-band. The isotropic sky assumption leads to a systematic under estimation of diffuse irradiance. Several correction methods have been proposed to take into account the anisotropic effects. However, their performance at a given site depends on the dominant local climate. In this work, it is shown that the local adaptation of shadow band correction methods results in a significant improvement in the diffuse irradiance measurement's accuracy. Nine well-known correction methods are implemented and tested (both in their original and locally adapted versions) for the Pampa Húmeda region of southeastern South America. In absence of local adaptation, only one of the pre-existing methods improves the simple isotropic model. All locally adapted versions perform similarly well and outperform significantly the original methods. A new model based on the parametrization of Battle's model is proposed. It provides the best performance compared to all locally adapted pre-existing models, under all-sky and discriminated sky conditions.