Now the idea is (with a desktop PC and the Delta 1010LT) to route 8 mics (Drums, Bass, two Guitars) into the analogue inputs and use the analogue outputs of the card as inputs to the Desk. Like this we could still use the desk for monitor mix and have 8 individual tracks recorded which can be remixed later and more tracks (voice or keys etc.) can be added.
Yes, this will work with two caveats.
Personally I think it would be easier to use the driver software to make your monitor signal. The way the driver software is set up, you can route any of the input pairs to any of the output pairs. You can also set channels 1+2 to be the "main mix". All the inputs and the computer sounds are routed to the main mix, but you can control each level separately in the software. Since you'll probably want output channels 1+2 as your main mix output, it might be easier to just set that up as a monitor mix as well by using the sliders in the driver software to set the monitoring level of each source. The drivers will mix these all down to 2 channels and put them out of channels 1+2. That way you just need a headphone amp/splitter and give everyone a set of headphones while recording.
Note that this is NOT Software monitoring, it's still Hardware monitoring. You're just using the driver software to tell the hardware what to do. It will add a small amount of latency to the monitoring signal (this is unavoidable), but the latency will be the same for each channel so it won't throw off your timing. The amount of latency is changeable in the driver software. I use 8ms without any trouble. You may have to bump it up a little bit higher if your computer doesn't run as smoothly as mine.
If you try to use the mixer to create the monitor mix, you'll have to constantly change the settings on the output channels back and forth between "main mix" and "input 1+2". That's why I think it will be easier to just use the driver software to create a monitor mix and leave it set up all the time.
Also, you'll have to use a roundabout method if you want to use the sliders on that mixer to make your mixes. Audacity does not currently support more than 2 output channels, but it can create multichannel wav files. If you really want to use the mixer, you can create an 8 channel wav file and figure out how to play it using a different piece of software. Then you can route those 8 channels through the mixer and record the 2 channel mixdown using Audacity. The upside is that you can use the sliders (which are more intuitive than software) and you'll be able to easily apply global effects (e.g. reverb) to all the tracks using the mixers effects send/return. The downside is that this will require some cable swapping and will put your music through one more step of digital -> analog -> digital conversion which will cause a (very small) loss of quality.
The alternative is to use Audacity to make all the mixes. This won't require cable swapping or rely on other pieces of software. But it will be more of a pain to apply effects like reverb to everything.
Even if you don't use the mixer for monitoring or mixing, it will still be very useful as a way to control recording levels and use hardware effects.