A few odd comments...

My motors run hot, depending on how they are being used. For example, a while ago I was cutting some 3D profiles which meant continuous running in the X direction (back and forth) for maybe 6 hours and occasional sideways steps in Y (at the end of each X pass). The Y motor became warm but not hot; the X motor was too hot to touch although it passed the "spit" test. I was a bit worried about this but doing some googling came up with the answer that stepper motors should be able to run like this. Given the capabilities of my machine, I could probably have reduced heating by reducing the current and still had enough torque available, but it didn't seem to be a big deal so I left it, and it did a number of those runs apparently without any harm to anything. Other cuts such as profile cutting out parts where both axes run maybe 50% of the time on average mean that neither motor gets that hot.

As for windings/current/torque, etc, it's all a bit complicated and some of it is not that intuitive, even if you've read the books. I'll try and hand-wave it, to see if that helps. For a given motor, torque is developed proportionally to the current through the winding(s). Double the current, double the torque. Torque also depends on how many turns of the winding the current goes through. Double the turns for a given current, double the torque. In effect, your 6-wire motors have two windings, each having a connection to the ends and one to the middle. For each coil, you could think of it as two windings in series, joined internally and with the common point brought out. So, based on my earlier comments, if you put some fixed current through the whole winding, you will get X amount of torque. Put the same current through half the winding (connect to end point and centre point) you will get X/2 torque. So why would you ever want to do that? The problem is that all the above is talking about "steady state" current, some time after you apply current and things have stabilised. When you first put a voltage across a coil, current will start to flow, but it takes a while to build up. It doesn't start flowing at full value immediately. How slow or fast it builds up depends on the inductance of the winding. Inductance is a bit like inertia - try pushing a trolley with a given force and it will start to accelerate. Put a load on the trolley, increasing its inertia, and the same force will make it accelerate more slowly. You might get up to the same speed, but it takes longer. Same with a stepper motor winding and its inductance. You are sending a series of pulses to the motor, and on each pulse you want the motor to turn to the next step. Low inductance coil/winding - current and therefore torque builds up quickly, motor moves to next step quickly. High inductance winding - current and therefore torque build up slowly and motor takes longer to respond. Put a load on the motor, and the low inductance version will be able to run faster as it builds up the torque on each current pulse to move the load more quickly; same load with a high-inductance motor and if you don't give it enough time on each pulse of current to build up enough torque to turn the load, it's going to sometimes fail to turn (missing steps) or in an extreme case, sit there vibrating a bit and not turning at all. That's why the recommendation is to use low-inductance motors. With a 6-wire motor, the best compromise is probably to use half the winding, which means only half the maximum torque, but also halving the inductance so you can keep up the stepping rate. With an 8-wire motor, you can choose to put each pair of windings in series (high inductance so limiting speed) or in parallel - low inductance so higher speeds, but you will need double the current capability from the power supply and driver as you will be putting the same current through each winding. The other way to get speed up is to increase the power supply voltage. Double the supply voltage and (roughly) you double the rate of rise of current. So you get to the final value more quickly on each pulse, so reach the torque needed to move the motor more quickly, so can run faster. If you read the data sheets, you might well see that a motor is rated at a nominal 4V but in a practical system, the power supply driving that motor will be at 65V. That is purely to make sure that the current rise is fast enough to develop max torque. The driver sorts out current limiting and so on to make sure that the motor is not overloaded - that's what the current setting on the DIP switches does, and you would typically set this to around the max motor current rating and ignore voltages. So the typical cheap drivers with their 24V supply will not allow you to run a motor as fast as you could with 48V or 70V. However, the higher voltage drivers and power supplies cost more.
As I say, that's all a bit hand-waving and there's plenty of theory that I've skipped or simplified, but having that kind of picture might help put some of the advice you've been given into context. It's all built on a sound technical background! Higher speeds come from low inductance motors driven by high voltages and drivers to suit; cheap kits have high inductance motors, low voltage supplies, and sometimes rather flaky drivers built down to a price. Tends to lead to disappointment all round!