Background: Breast cancer is the main cause of cancer mortality among women. Green spaces have been recently associated with reduced cancer mortality among women. Mechanisms explaining the beneficial effect of green spaces include increased levels of physical activity and reduced exposure to air pollution, which have been both associated with cancer development. Objectives: To investigate the associations between presence of urban green areas, presence of agricultural areas and surrounding greenness and risk of breast cancer, and to assess whether these associations are mediated by physical activity and/or air pollution levels. Methods: We geocoded the current residence of 1129 breast cancer cases and 1619 controls recruited between 2008 and 2013 in ten provinces of Spain, as part of the MCC-Spain study. We assigned different indicators of exposure to green spaces in a buffer of 300 m, and in nested buffers of 100 m and 500 m around the residence: presence of urban green areas according to Urban Atlas, presence of agricultural areas according to CORINE Land Cover 2006, and surrounding greenness according to the average of the Normalized Difference Vegetation Index. We used logistic mixed-effects regression models with a random effect for hospital adjusting for potential confounders. We explored the effect of several potential effect modifiers. We assessed mediation effect by physical activity and levels of air pollution. Results: Presence of urban green areas was associated with reduced risk of breast cancer after adjusting for age, socio-economic status at individual and at area level, education, and number of children [OR (95%CI) = 0.65 (0.49-0.86)]. There was evidence of a linear trend between distance to urban green areas and risk of breast cancer. On the contrary, presence of agricultural areas and surrounding greenness were associated with increased risk of breast cancer [adjusted OR (95%CI) = 1.33 (1.07-1.65) and adjusted OR (95%CI) = 1.27 (0.92-1.77), respectively]. None of the associations observed were mediated by levels of physical activity or levels or air pollution. Conclusions: The association between green spaces and risk of breast cancer is dependent on land-use. The confirmation of these results in other settings and the study of potential mechanisms for the associations observed are needed to advance the understanding on the potential effects of green spaces on health.