First, Would like to understand which application you’re automating. its Web, Desktop, Citrix?
If you’re able to select button using indicate on screen, then this button will have a selector, which contains all the unique attributes of that particular button. You can remove some attribute and make it generic.
Regarding, Click Text and Click Image - Click Text will first find text on the screen then it will click on it. Click image will try to match image which you have captured during element selection and then it will click. Both can be used… click image might not work if resolution or size of button is changed.