Albert,
I'm using similar SBR20 rails for my long axis (X on my machine) rails and SBR16 on the gantry. I've been happy with them so far. The accuracy of your machine is going to depend largely on how straight all the rails are. Then it's down to to how rigid everything is and how well you can align the beast once it's built. Keep in mind how you are going to do that during the design phase.

The 1000mm rails are going to allow you a maximum of around 800mm travel if you allow about 200mm for the gantry feet. The ballscrew might be the limiting factor there as the maximum possible travel on the screw will be the length of the actual threaded part minus the length of the ballnut. If you want to keep the length of the overall machine down then you will have to drive it with a belt and pulleys.

NOTE: If you run the ballnut off the screw all the balls fall out. You wouldn't be the first person to get caught out by that one. If you need to run it off to turn it round it can be done quite easily though.

650mm for the gantry mounted rails and ballscrew will get you less than 500mm travel and depends either on the ballscrew and nut as described above or the width of the Z axis assembly (mine is 160mm wide which is a tight fit for all the bits but is a standard size for aluminium bar stock) but with a single screw on the long axis you cannot expect to drive too wide a gantry anyway.

If you really are tight for space then I think you could keep the machine in a footprint no larger than the length of the rails/ballscrews using the parts you have bought but it would be easier not to and you will lose some of the available travel, especially on the long axis. You are clearly restricted by the sizes of the aluminium profile you have available which is going to affect the decisions made. This doesn't mean you cannot build a machine capable of useful work out of what you have but some compromises will be required. Based on what you've said, the construction of such a machine will have served it's purpose anyway. Once you have proved your own ability to make a machine that actually works you can then decide whether to commit the funds to doing what almost everyone else does. make a bigger one!

Kit