However, if you do the math, you'll see that even the earliest possible galaxies (redshift = z = 10 (or less)) don't get visible light redshifted into the submillimeter (where Herschel looks). This is because the earliest galaxies are very dusty, and this dust obscures most of the visible light coming from the stars. However, to obscure the light, the dust is absorbing the energy, heating the dust up (to about 35K give or take) and this dust re-radiates at like 10-100 microns or so, which is redshifted into the submillimeter (200-600 microns) that Herschel looks at.

I work on a precursor to the SPIRE instrument (a major instrument on Herschel, called BLAST which flew on a high-altitude balloon (you need to get about the atmosphere to look in the submillimeter and balloons are way cheaper than sattelites).

And it's not really a Spitzer replacement, as it's looking at mostly longer wavelengths than Spitzer. It complements both. Virtually all different wavelengths are worth looking at as they all tells us unique things.

Herschel, mostly with the SPIRE instrument, will be looking at the earliest star forming galaxies. A nearly identical instrument to SPIRE was flown on a high-altitude balloon, BLAST, which got Slashdot coverage as well.

Disclaimer: I work on the BLAST project.