There's so much BS about cooling, and so many things that people simply don't understand without a thorough education in REAL thermodynamics and fluid dynamics.
Flow is hugely important. Why? Because mass flow rate dictates how quickly you suck heat out of system. This is dependent on many variables, but in general more flow is better. A well designed system will work at a temperature matched to a flow rate that maximizes heat extraction.
There's more than one kind of pressure. There's static, and dynamic. Cooling systems don't give a flying F about static pressure - the pressure on the radiator cap. Because it's the same everywhere. Dynamic pressure is different. This is pressure that results from a restriction. Think putting your thumb over the end of a garden hose. So having coolant passage restrictions that cause a reduction in flow will cause a spike in pressure. If these are well-placed, they can help mitigate local hot spots that might otherwise flash boil at lower pressures. I'd argue that if this is necessary, then the system FLOW should be increased to fix it first - but that's not always a choice OEMs can make. They have to package that cooling system somehow.
Lastly: shedding heat. This is where the size of the radiator comes into play. While there's no such thing as having too much flow, you CAN have too little surface area to shed the heat out of the mass of coolant. This is where #'s of cores come into play. A very high flow radiator can work well, but once the coolant and radiator are heat soaked, it simply may not be possible to shed the heat off the limited space fast enough. It becomes a balancing act. The REAL art is getting big flow while maintaining the surface area needed. This is why good parts cost money - it takes engineering. Just like a well ported head versus bubba with a dremel.
As far as water/glycol/wonder-fluids: the name of the game is boiling point and SPECIFIC HEAT CAPACITY. Water has a very high specific heat capacity. Meaning it takes a lot of energy to heat it up. The problem is that it also boils at a relatively low temperature, and once it does - it stops absorbing heat so well. This is where glycols and other additives come into play. They help augment the boiling point while retaining some of the properties of water, but also contain other additives which are good at absorbing these funny things called ions that would otherwise react with the various metals in our motors. You can get away with reducing specific heat capacity, but it REQUIRES adding mass (more volume) to the system to maintain equilibrium. If you're working with an OE system at OE power levels, you can probably get away with wonderfluids and no water, but if you have a tiny radiator with a 2,000hp mill behind it... don't count on it working so well.
This is where STATIC pressure comes in. When the static pressure is raised, it increases the boiling point of the water so that it can continue to absorb energy/heat. Pascal's law tells us that pressure on a fluid is felt anywhere that fluid contacts the vessel. This means your cooling system doesn't care if it's at 0psi, or 120psi - but your components do. Whether due to the difference in boiling point or the mechanical strain created by the pressure.
So in the end, you need ENOUGH coolant to absorb the heat generated while at the fastest flow rate that can be reasonably achieved. Then the radiator MUST be sized to shed that amount of heat at a faster rate than the coolant is sucking it out. This is done to ensure that the motor can always be cooled below the thermostatic temperature and that it never runs away. Coolant mass can be added to suck more heat out, but all it does is delay the inevitable boil-over because the radiator simply MUST shed as much or more heat than the engine can produce.
As far as thermostats: they can function as flow restrictors and cause localized pressure increases to stop localized boiling. A properly engineered system wouldn't need this part of the function, however.
Engine operating temperature is a function of many things. The cylinder wall wear we see in conjunction with cold engines has to do with the gasoline 'washing down' the walls and eliminating the lubrication for the piston rings. A direct injection motor wouldn't care if it's at 120F, 20F, or 205F so long as the combustion is efficient - the cylinder walls will never see gasoline vapors or liquid. So the induction method and type is important. For a carb, we want a hotter motor to help vaporize the fuel - to a limit. Too much heat starts to reduce charge density and can exacerbate pre-ignition. A fuel injection motor, well it depends. Port fuel injection will tend to atomize well due to turbulence, while a TBI relies on the same mechanism as a carb. Sequential injection offers the benefits of port injection with the added benefit of firing in time with valve events to maximize the vaporization and minimize liquid fuel droplets. This would allow the engine to run at a lower temperature without the adverse effects.
BUT, it can become less consistent because the engine is running closer to ambient temperatures which can vary wildly.
So, in the end, it's just like anything else: build it for the job and don't get caught up in the marketing hype. A bigger radiator never hurts, but the wrong coolant or too high or low of a stat, or too high or low of a pressure can cause adverse results unless they're well thought out and the system is designed to operate with them.
