The defining metric for any wireless communication network is the maximum reliable data rate, also known as capacity. Before any data bit can be communicated over a wireless channel, information about the network state such as connectivity, channel statistics, or channel gains, is required. Receiver nodes feed such information back to other wireless nodes using feedback channels and feedback mechanisms. Considering the role of feedback makes the characterization of the network capacity a daunting theoretical task. As a result, researchers have overwhelmingly simplified the feedback channels to come up with tractable models. Specifically, it has commonly been assumed that feedback channel has infinite capacity and has no delay. While these assumptions could be justified for small, static, or slow-fading networks, they are not viable in the context of large-scale, mobile, or fast-fading wireless networks. In fact, feedback channel is low-rate, unreliable, scarce, and is subject to delay. The recent dramatic increase in wireless data traffic, caused by the success of online media streaming and the proliferation of smart phones and tablets, obliges researchers to understand the capacity of large-scale mobile networks. Thus, given the limited, scarce nature of feedback channels, future progress in wireless data communications crucially depends on a deeper understanding of the impact of feedback on the capacity of wireless networks. In this work, we aim to adjust the assumptions on feedback channels in a way to open doors to better understanding of real world, large-scale wireless networks. In particular, wireless networks are considered with rate-limited feedback links, with outdated channel state information, and with local views of the network state.