The study of restorative environments has gained increasing attention in environmental psychology because of its implications for health and well being. Traditionally, this research topic has been addressed through the comparison between the restorative potential of natural and built environments, showing a higher impact of the former on well being. Conversely, the potential for restoration of different typologies of nature has remained a somewhat undeveloped issue. The role of biodiversity seems to be an important aspect to be considered in this respect. In addition, a consistent amount of literature has focused on the outcome of being in contact with restorative environments, while less attention has been paid to the process leading to restoration. In this study, we selected five typologies of urban green spaces in Italy, ranging from a minimum of biodiversity and a maximum of man-made elements to a maximum of biodiversity and a minimum of manmade elements. The environments were an urban plaza with green elements, an urban park, a pinewood, a botanic garden, a peri-urban natural protected area. A convenience sample (N = 125) was contacted in the city of Padua (25 participants for each of the five different typologies). A questionnaire focusing on people’s experience in the environment (length and frequency of visits, activities performed, perceived restorativeness, affective qualities of the place, perceived well being during and after the visits) was administered. The five typologies of green spaces considered were compared for their perceived restorative properties (e.g., being away, fascination, extent, and compatibility). Also, the relationship between individual exposure to green spaces (activities performed within the green spaces, duration and frequency of visits) and perceived well being, as well as the mediating role of perceived restorativeness and affective qualities, were tested. Results showed that the perceived restorative properties are higher in the peri-urban green areas, and significantly increasing as a function of biodiversity levels in the environment. Moreover, the activities performed in the environment impacted both perceived restorativeness and respondents’ well being. Finally, frequency and duration of visits positively predicted self-reported well being. As expected, a significant mediating role of both perceived restorativeness and affective qualities upon the relationship between duration and frequency of visits and individual well being was detected. The theoretical implications in the analysis of the process leading to restoration are discussed, and potential guidelines in view of a more healthy management of everyday urban and peri-urban natural environments are envisaged.