If you use a Limiter correctly, it accomplishes the normalizing step, and it is dynamics processing. Also, a graphic equalizer is a very blunt tool. You’d be better off learning to use parametric equalizers if you’ve got specific problems.
Jester’s guide is good. Read it, then read it again.
For my take on it, what I usually did was:
Mix → Pristine/Clean Channel Strip → Colored Gain Stage → Limiter → mp3 encoder.
My favorite clean channel strip is the one that comes with Pro Tools that runs the same algorithms as Avid’s top-end consoles. It’s very clean and easy to use. I like it because you can basically draw in your changes and fine-tune them with your ears.
Typically, I’d use the Dynamics/Expander section to very lightly tame volume peaks (with a limit on how much compression it would apply, a great feature) and a fairly aggressive (to my ears) expander to give over-compressed tracks a bit more room to breathe.
The EQing I’d apply changed based on the gear use…but most of the time I’d use the EQ for a gentle HF rolloff, since most digital mixing kind of lets noise build up there, especially if the whole signal chain was digital. Other than that, I just listened for problems and got rid of them with the parametric EQ, usually with some automation. Basically, this let me not tweak EQs as much while I was spinning since I could do it more subtly after the fact. (note: we’re not talking about bass cutting or other EQ effects…we’re talking about balancing different parts of the mix). I also tended to very gently high-pass out some of the sub-bass as well. I figured my mixes would mostly be listened to on headphones/earbuds or computer speakers that didn’t have impressive subs anyway. So, having a lot of sub bass took up headroom but didn’t really get me anything.
I’d try them in different orders, but IIRC it usually went filter → dynamics → EQ if the strip was capable of it (or if I was using separate plugins).
After that, I’d use an analog-emulating plugin as a gain stage.
Note: With an analog mixer, I was feeding my sound card with a nice, big analog signal (not clipping, but not all that quiet either). With a digital mixer (Traktor or Live internal), it was pure digital signal straight onto the drive. Once it was in the DAW software, though, it wasn’t uncommon for me to set the software gain so the mix peaked at like -18dBFS so I knew I had plenty of headroom to work with. As long as the input is good, the internal representation being quiet isn’t a big deal…modern DAWs use 32-bit float numbers, and the noise floor is so incredibly quiet it just doesn’t matter.
So, that gain stage was an emulation of some piece of gear I was enamored with at the time. At home, it was usually an SSL 4000’s Master Bus Compressor because those plugins are cheap and plentiful and usually pretty good. Sometimes, It’d be a Urei 1176 or Neve 33609 compressor. When I’ve been lucky enough to work or study in studios with my yearly income in plugins, I’d use things like the UA API preamp models. This is analogous to the saturator that Jester was talking about in his guide, except I got to exercise my gear wankery a little without having to shell out thousands of dollars for outboard gear.
That ran into a clean, digital brick-wall limiter, possibly with a clean gain stage after the analog-emulated one depending on how aggressive the model was. The limiter was initially set so the mix would not go above about -0.3 dBFS, and I played with the 3 gain stages (output from the channel strip, the analog-emulated one, and sometimes a clean digital one if the limiter didn’t have one of its own) until it sounded good on whatever I considered my “reference” monitoring (usually my MDR-7506s, preferably the Genelec 2.1 and yamaha NS-10 monitors I used to have access to). I usually had a plugin right after the limiter that would check for “intersample modulation distortion”, and I added headroom if that light ever tripped, making sure the limiter never gave more than a couple dB of gain reduction.
Often, I was “done” then. Sometimes, I’d listen in the car, on my phone, etc. to see how it sounded there, but I usually didn’t bother.
Then the mp3 encoder.
What this did was create a mix that breathed (actually had dynamic range), sounded good most places, let the loud parts feel/seem just a bit louder, and at least could use all the dynamic range that digital has (even though it didn’t matter except for reverb tails when nothing else is happening).
I’m not sure how that applies directly to Cool Edit Pro, but it’s a slightly different take. The details of what plugins or fake gear I used aren’t important except to say that I kind of prefer gear emulations to more general saturators. The only saturator I ever really liked was this multiband thing that I usually used to “build” guitar amps…you could just set it really light (compared to a guitar amp)…but it was so easy to overdo, I think I only did that once.