Installation This explains how to do different MOOSE installs, from single-click for Linux and Mac environments, to CygWin for Windows, to the full compile source tree.
GUI This explains how to use the MOOSE GUI. There are a number of click-and-run demos on the MOOSE GUI launch interface, but the GUI also supports more advanced features for loading, running and editing models. The GUI documentation leads you through these operations.
Rdesigneur This section is the way to go to learn how to build detailed single neuron models, chemical reaction-diffusion models, or models with both. The very compact Python code using the Rdesigneur tool is the best way to build, run and display demo and production models.
Tutorials A set of tutorials for teaching a range of topics, including integrate-and-fire networks, chemical bistables, and oscillators. Has stand-alone graphics and the Python scripts are meant to tinker with.
Snippets A diverse collection of snippets: (mostly) short code examples illustrating different features of MOOSE
MOOSE Cookbook The MOOSE Cookbook contains recipes showing you how to do specific tasks in MOOSE
MOOSE  quickstart Tutorial on using the Python scripting features in MOOSE
MOOSE user reference Automatically generated documentation for MOOSE built-in functions and MOOSE classes, as accessed through Python.
Source code reference Documentation for developers, plus Doxygen-generated code documentation.
If you want to develop MOOSE, go here. This page includes a Programmer's Guide, API documentation, and a Design Document. Further, MOOSE code is internally documented using Doxygen. This provides function and class-level information for developers. On this site you will find the current Doxygen-generated code documentation.