Whether you decide to build or buy, the flight controls for your sim ideally should faithfully mimic the feel of actual flight controls. This is no small feat. Actual flight controls display some rather complicated behavior, and it may not be feasible to mimic it all. Nonetheless, let's take a look at what real flight controls are like. This will give you a firm background when you check out the commercially available controls or plan to build your own.
Aircraft flight controls are designed to operate smoothly and be reasonably sensitive. They don't always operate as smoothly as they are designed to be, there is friction in any system after all, but you won't, with luck, find a flight worthy aircraft with sticky, jumpy flight controls. By sensitive, I mean that if you move the yoke a tiny bit, the aircraft will respond, even if just a tiny bit. There is no reason that a good set of sim flight controls shouldn't function this way. If they don't, keep shopping.
Now things start to get a bit more complicated. Flight controls both feel different and act differently as flight conditions change. Changing airspeed is a big factor. As the airspeed increases, flight controls (at least in non-fly-by-wire A/C) become more effective and feel increasingly stiff. In a non-fly-by-wire aircraft the range of this effect is dramatic. While taxiing, flight controls are pretty much ineffective. At the top of an aircraft's operating airspeed, you can get yourself in trouble with inappropriately large control movements. Modern fly-by-wire airliners moderate these effects with flight control systems. In particular, the "control laws", basically the algorithms coded into the flight computers, disallow control movements that would damage the aircraft. Smaller, and older, aircraft still experience these effects in full.
Flight controls can do a Jeckle and Hyde routine when the aircraft stalls. Airflow over the wings shifts from laminar (smooth) to turbulent. With small aircraft, this can mean that the horizontal stabilizer and rudder are surrounded by the turbulence generated by the stalling wing. Their effectiveness drops and the controls feel different. In extreme cases there may be no control, though contemporary aircraft are designed to mitigate the effect. Burt Rutan's canard designs, like the Varieze and Long EZ are designed such that the canard stalls first, causing the nose to drop a bit therein keeping the main wing away from a stall. One approach used in many small general aviation aircraft is to twist the wings slightly. At the root, the wings present a slightly greater angle of attack than the outer portions with the ailerons. This means the wing roots stall first, leaving the ailerons in laminar airflow so they are still effective in controlling the aircraft. If the whole wing stalled at the same time, the pilot would have no roll control. Likely one or the other wing would drop, and the aircraft would loose a fair bit of altitude. As most unintentional stalls occur during landings and takeoffs, this could make for a really bad day. Ideally, the stall warning indicator should be buzzing away, but in any case, the shift in feel of the controls presents the pilot with a cue to how the aircraft is flying.
There are changes in the feel of the flight controls that occur more slowly as a result of more mundane factors, for example shifts in an aircraft's center of gravity. Depending on the aircraft, fuel usage causes such a shift. Generally, most aircraft have their fuel tanks in the wings where fuel usage has minimal impact of center of gravity. But some very long range aircraft, strategic bombers for example, have additional fuel in the fuselage. In commercial airliners the center of gravity shifts when the seatbelt light goes out and a third of the passengers sprint for the bathroom. As the center of gravity shifts, the pressure required on the yoke for straight and level flight changes. Of course, the pilot trims this out, but the changes in pressure are an indication to the pilot of what's happening with his aircraft. Generally an experienced pilot will anticipate these changes. It's the unexpected ones that are important. A rapid shift in yoke pressure while flying cargo is quite possibly a clue that something unhappy is about to come down.
These examples are the very barest of introductions to aircraft handling. If you're interested in pursuing the subject a bit further, a great next step is reading the book Stick and Rudder, an Explanation of the Art of Flying by Wolfgang Langewiesche. This is the classic. Because it was first published in 1944 you might think it's outdated. Let me assure you the physics of flying has not changed in the past several decades. Langewiesche's explanations are intuitive and easily understood. The fact that the book is still in print says a lot. I also recommend David Thurston's book, Design for Flying. Of course, the best way to learn about aircraft handling is first hand.