Kj -
Look at your 34465a specs. Those "Low DC current" specs are not hard specs, only "typical" - that means there is no max / min specified uncertainty limit. Consider those a ballpark measure only, without any guaranteed accuracy.
A calibration will come in many levels - a basic cal will be that the instrument is just checked to be within spec (and that can be anywhere with published spec). Calibration with data (aka "Z540") will come with actual measurement data.
You can also get a calibration with adjustment - which you may or may not want. If you're building a long drift history on your equipment, you normally DON'T want any adjustment done IF your device is still within specification - and you'll correct your measures for drift yourself. If you're using a meter say on a production line test and you sort of want those raw data measures to nearly match other meters on the line without any other drift correction - then you request an adjustment to be "within X ppm" of some nominal value, on whatever range you're interested in. But normally this won't be done unless you request - and usually pay - for that extra service. If the meter goes in for cal / repair and a measuring range is out of spec: Depending on the situation this becomes an "adjustment" or "repair" - it just depends on what is wrong, and how long will it take to get the instrument pulled back into spec.
Your meter is perfectly fine and within specification. You're measures should be within the stated uncertainty for each range. It is important to always be aware of the equipment limitations and absolute value uncertainty when going after low ppm measures - and take into account drift and expected uncertainty as you build up a drift history record unique to each piece of equipment.