In the article there are presented genesis, basic assumptions and pedagogical results of IT School Program based on personalized IT virtual learning system (PITLS) designed and implemented for secondary school students and teachers in Poland, run from November 2012 by Warsaw School of Computer Science. Author explains the aims of the Program, its theoretical key layouts such as usefulness, partnership, networked learning environment, diversified and high professional level of materials, interactivity mechanisms, personalization mechanisms, built-in incentive mechanisms, and automation of selected elements of the educational process and system data analysis. Some examples of the virtual educational tools and techniques used within the Program are presented. Great emphasis is put to explain the pedagogical and statistical results of the Program based on author’s own research surveys done within three years of Program duration on population of teachers and students taking part in IT School Program and on the basis of big data analysis results generated by PITLS. Computer assisted web interview (CAWI) research method was used as the main source of getting research material to be analyzed. Finally the main future directions of IT School Program development are shortly discussed. In conclusions author will among others try to answer the question what is the reason of the phenomena that almost 70 000 students from over 500 secondary schools registered in the Program within 2,5 years since its starting and have performed altogether almost 400 000 online courses and how it corresponds to achieving pedagogical aims of the Program. The openness of the program places it within Open Educational Recourses ideas, massive number of participants involved and number of courses performed resembles MOOCs. But the Program has a unique networked virtual organization (NVO) structure, thus it is rather Networked Virtual School (NVS), which has characteristics going beyond what is described today as OER or MOOC. Although no formal comparison study is made, one of the aims of the paper is to prove that NVS is much more effective in achieving expected pedagogical results then OERs and MOOCs.