Purpose of Review: Natural archives are imprinted with signs of the past variability of some aerosol species in connection to major climate changes. In certain cases, it is possible to use these paleo-observations as a quantitative tool for benchmarking climate model simulations. Where are we on the path to use observations and models in connection to define an envelope on aerosol feedback onto climate? Recent Findings: On glacial-interglacial time scales, the major advances in our understanding refer to mineral dust, in terms of quantifying its global mass budget, as well as in estimating its direct impacts on the atmospheric radiation budget and indirect impacts on the oceanic carbon cycle. Summary: Even in the case of dust, major uncertainties persist. More detailed observational studies and model intercomparison experiments such as in the Paleoclimate Modelling Intercomparison Project phase 4 will be critical in advancing the field. The inclusion of new processes such as cloud feedbacks and studies focusing on other aerosol species are also envisaged.