Long answer: when sending out beams when "locking" in air/air mode the radar uses two return intervals, one "long" and one "short". It sends out a beam and waits for the return on a specific time, either the long return interval or the short one. At first the intervals are set close to the aircraft, where the short one is the time it takes the beam to travel the closest possible distance and back and the long one is the time it takes the beam to travel to a slightly longer range and back. If there is no return signal in, for example the short range interval, the intervals are increased proportionally until both intervals has a return signal. If there is a return signal in both timeframes the interval is frozen. This way the air is longitudinally scanned until both intervals has return on a somewhat common distance, and that distance is where the target is assumed to be. There is no information about the bearing or pitch to the target when "locking", and the wording is a bit confusing. The point of why they designed it this way is to minimize interference, esp ground interference.