Maybe you had a different sort of camera. Noise on my DSLR is more about the ISO. I can leave the shutter open all day and not have any noise, as long as the ISO is low (under 800).
The shortest exposure I would ever work with is maybe 15 seconds.
I used a few. One was a peltier cooled 1k x 1k CCD. The other was a little smaller and was liquid nitrogen cooled.
Thermal noise from the telescope being at ~10C on observing nights drowns out a lot of sources.
To get the really faint stuff you cool your CCD to get the IR noise as low as possible and take dark field images of the inside of the camera with the shutter closed to measure whatever IR is left. Every 30 minutes or so you take ~10 or so dark fields, average them, and then you subtract the dark fields from the next set of images you take so you properly zero them out.
With a CCD that sensitive, the really bright stuff will cause some of the pixels to bleed over into others, so you have to keep the exposures short. This means taking a lot of them.
And then there's taking away the dirt from the optics, so you use flat field images to correct for that.
I haven't done it for a long time, but as I recall, the procedure is something like:
20 flats (taken of the bright sky just around sunset)
20 darks
Then wait for your target to come up
10 R
10 G
10 B
10 darks
10 R
10 G
10 B
10 darks
and you repeat until you have the area you want mapped out. Then in post processing you average the set of 20 flats and each set of 10 darks, then subtract the darks from the following 30 RGB shots (or whatever filter you used). You then divide all of the RGB shots by the flat field to get ride of the dust on the optics. Then you align the images, stitch them all together and colorize.
The last observing run I did was a map of the Pleiades with a 42cm scope. I can't for the life of me find the image though