The evolution of galaxies is tightly linked to processes that occur in the circum-galactic medium (CGM). Unfortunately, most of the baryons in the CGM are too faint to be easily detected. At high-z, we view glimpses of dark baryonic halos through quasar absorption lines, or cooling-radiation emitted as Ly-alpha. However, a direct connection with the stellar growth of massive galaxies has long remained missing, because we had yet to identify the ultimate reservoir of halo gas that has sufficient mass to fuel widespread star-formation, namely the cold molecular gas (~10-100 K). I will present sensitive low-surface-brightness CO and [CI] observations of the massive Spiderweb Galaxy (z=2), which reveal a very extended (~70 kpc) reservoir of cold molecular star-forming gas. This cold CGM has a carbon abundance and excitation conditions similar to the ISM in starforming galaxies, implying that massive high-z galaxies grown not directly through accretion of pristine gas from the Cosmic Web, but from recycled gas in the CGM. Based on this and several other examples, I will explain the technical challenges involved in lighting up these hitherto hidden reservoirs of cold molecular CGM. This leads to the question of how much molecular gas in the Universe we are still missing? I will show how low-surface-brightness observations with future radio interferometers, with particular emphasis on the Next-Generation VLA, promise to revolutionize our view of the molecular Universe.