No, and the more correct terminology is 'pseudo random'. A good PRN generator will produce, on subsequent calls, a sequence of numbers that have many of the properties of truly random numbers such as statistical independence and an autocorelation function which is 0 (except at 0 lag). But while a modern generator will produce very long runs, eventually the sequence will repeat.

The sequences are also pseudo random in the sense that you can get the same sequence every time by 'seeding' the generator with the same initial state. The seed goes in the parentheses e.g. RAND(seed) at the first call. RAND will produce the same set of 1000 numbers for each sequence you generate in which the first call uses the same seed.