All the items used for measuring one construct like satisfaction in your case, should be measured on the same scale. For example, if you are measuring one item on 3-point scale and other on 5-point scale, then its a problem. But in your case I see that you are using a 5-point scale for all items, which I think should not be a problem.

There is absolutely no problem with using scales with different categories or of different lengths in the same research project. However there are three important considerations.

1 scale length eg 3 or 5 point scales

you should perhaps from a preliminary study make sure that the scale length is meaningful to respondents. Eg don’t use a long scale when the likelihood is is the respondents make binary choices. Too long a scale will be difficult to use. We normally recommend 7 points as the maximum. However shorter scales will give less refined results if the respondents could easily discriminate say 5 or 7 points.

2 these scales cannot be assumed to be linear. That is to say the gaps between each pair of points may not be the same. This means that you should only use non parametric statistics when analysing them

3 if you wish to draw conclusions about the relationships between the scales ie about answers to different questions then if you want correlation the results are likely to be less significant if the scale lengths are not the same. When showing differences then chi square and other such tests can easily work with say a 2 point scale and even a 7 point scale.

It is okay. Many scales that are widely used have different anchor points for items for a given construct. See Lyubomirsky and Lepper's scale on subjective wellbeing for instance.