This paper presents a detailed statistical channel model for millimeter wave (mmW) frequencies (28 and 73 GHz) in New York City, derived from real-world measurements, and evaluates the potential capacity of mmW micro- and picocellular networks in dense urban environments. The models are used to assess the performance of mmW systems in terms of path loss, spatial clusters, angular dispersion, and outage probabilities. The results show that even in highly non-line-of-sight (NLOS) environments, strong signals can be detected up to 200 meters from potential cell sites, potentially supporting spatial multiplexing. The models predict that mmW systems can offer an order of magnitude increase in capacity over current 4G networks without increasing cell density. The paper also introduces a three-state model for link conditions (LOS, NLOS, and outage) and provides statistical models for key channel parameters, including the number of clusters, cluster angular spread, and path loss. The results indicate that mmW systems can achieve significant spatial multiplexing gains, and that system performance is robust to outages, provided they are not significantly worse than those observed in the NYC measurements. The paper also discusses the potential of mmW systems for future Beyond 4G and 5G networks, highlighting the benefits of high-dimensional antenna arrays and beamforming. The models are validated against existing 3GPP models and show that mmW systems can achieve significant gains in capacity and performance. The paper concludes that mmW systems have the potential to provide a significant increase in capacity and performance for future wireless networks.This paper presents a detailed statistical channel model for millimeter wave (mmW) frequencies (28 and 73 GHz) in New York City, derived from real-world measurements, and evaluates the potential capacity of mmW micro- and picocellular networks in dense urban environments. The models are used to assess the performance of mmW systems in terms of path loss, spatial clusters, angular dispersion, and outage probabilities. The results show that even in highly non-line-of-sight (NLOS) environments, strong signals can be detected up to 200 meters from potential cell sites, potentially supporting spatial multiplexing. The models predict that mmW systems can offer an order of magnitude increase in capacity over current 4G networks without increasing cell density. The paper also introduces a three-state model for link conditions (LOS, NLOS, and outage) and provides statistical models for key channel parameters, including the number of clusters, cluster angular spread, and path loss. The results indicate that mmW systems can achieve significant spatial multiplexing gains, and that system performance is robust to outages, provided they are not significantly worse than those observed in the NYC measurements. The paper also discusses the potential of mmW systems for future Beyond 4G and 5G networks, highlighting the benefits of high-dimensional antenna arrays and beamforming. The models are validated against existing 3GPP models and show that mmW systems can achieve significant gains in capacity and performance. The paper concludes that mmW systems have the potential to provide a significant increase in capacity and performance for future wireless networks.